I wrote this piece because two different people asked me in the same week if there were an article such as this… An excerpt:
Although Friedman’s analysis sounds perfectly reasonable, and the epitome of “scientific,” Mises thought it was a seductive trap for economists. For a quick illustration of the difference in perspectives, let me relay an example from my teaching experience.
It was a principles of microeconomics class, and we were using the (excellent) textbook by Gwartney, Stroup, et al. In the first chapter they have a list of several guideposts or principles of the economic way of thinking. As I recall, these are items such as, “People respond to incentives,” and “There are always tradeoffs.” These were noncontroversial things that every economist would agree were important for getting undergrads to “think like an economist.”
However, the one guidepost that stuck out like a sore thumb announced, “To be scientific, an economic theory must make testable predictions.” I explained to the class that even though this was a popular view among professional economists, it was not one that I shared. I explained that everything we would learn the entire semester from the Gwartney et al. textbook would not yield testable predictions. On the contrary, I would simply teach them a framework with which they could interpret the world. The students would have to decide whether the framework was useful, but ultimately their decision wouldn’t boil down to, “Did these tools of supply and demand make good predictions?”
After I went through my spiel, one of the students made the excellent observation that not a single one of the other guideposts was a testable prediction. He was right! For example, how could someone test the claim that, “People respond to incentives”? I could say to a person, “I’ll give you $20 if you cut off your big toe.” Regardless of what happens, my claim is safe and secure. If the person doesn’t cut off his big toe, it just shows that I didn’t offer him a big enough incentive.